From aa2a1fe08f6d603b47b8577e704775e85261f877 Mon Sep 17 00:00:00 2001 From: Ilker Manap Date: Wed, 16 Sep 2026 23:38:09 +0200 Subject: [PATCH] Scale the settling window in the e2e tests instead of leaving them failing MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 4f9fef2 was pushed with test/e2e red, the second time in this session I printed a success line from an unconditional echo rather than reading the output above it. The check now gates the commit rather than narrating it. The failure was the change working as designed: a minute of settling against tests that wait fifteen seconds. Those tests register the whole fleet up front, so the window is scaled alongside every other interval there rather than switched off — the settling behaviour is worth exercising end to end, not excluding from it. Co-Authored-By: Claude Opus 5 (1M context) --- test/e2e/phase3_test.go | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/test/e2e/phase3_test.go b/test/e2e/phase3_test.go index 1ad39552..dfb18332 100644 --- a/test/e2e/phase3_test.go +++ b/test/e2e/phase3_test.go @@ -180,6 +180,10 @@ func startControlPlane(t *testing.T, nodes int, minReplicas int) *controlHarness ccfg.Reconciler.MaxBackoff = 500 * time.Millisecond ccfg.Reconciler.UnreachableAfter = 300 * time.Millisecond ccfg.Reconciler.LostAfter = 600 * time.Millisecond + // Scaled like every other interval here rather than disabled, so the + // settling window is part of what these tests exercise: the whole fleet is + // registered up front, and placement should resume once it holds steady. + ccfg.Reconciler.SettleFor = 400 * time.Millisecond ccfg.PublishInterval = 100 * time.Millisecond ctl := control.New(ccfg, cat, disp, rt, slog.New(slog.DiscardHandler))